2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering

th

QUANTIFICATION TOOLS IN INHERENTLY SAFER DESIGN

Rifai M.C.1, Bandoni A1.,Tonelli S.M.1*
1

Planta Piloto de Ingeniera Qumica PLAPIQUI (CONICET)

Abstract. In the last years, more rigorous environmental regulations and public concern about technological
risks has led chemical plant designers to pay special attention to all aspects related to safety. This matter is two fold. First, the traditional method involves the use of safety devices or control procedures added in the last design stages to deal with the identified hazards. On the other hand, in inherent safety, the risks associated to process materials and operations are reduced or eliminated permanently. The emphasis is placed on avoiding or limiting hazard sources rather than introducing add-on protective barriers to control them. The interest in inherently safer design has grown along the nineties because this methodology seems likely to be the most cost-efficient strategy to loss prevention. However, in spite of being an attractive approach, it is not widely used yet. Some reasons for this may be the fact that the industry is still more familiarized with conventional safety strategies and many of the proposed methodologies are not so simple to be implemented easily. Nevertheless, the lack of systematic tools to apply inherent safety principles is perhaps the most important cause. Many papers have recently been published attempting to deal with this drawback (Gentile et al. (2003), Khan et al. (2003), Gupta and Edwards (2003), Palaniappan et al. (2002a; 2002b)). In this context, this paper presents a state-of-the-art review of the existing indexes to evaluate and compare different alternatives in inherently safer design (INSET Toolkit, Overall Inherent Safety Index, Inherent Safety Index, Fuzzy Based Inherent Safety Index, Integrated Inherent Safety Index, etc.). The extent of their applications is discussed as well as their strengths and weaknesses using a literature case study. To provide a common basis for this analysis, the concepts involved in each index are applied to select safer process alternatives for acrylic acid production.

Keywords: Process Safety, Inherent Safety, Inherently Safer Design.

1. Introduction
For many years, conventional design of chemical plants has been primarily driven by issues related to economics and engineering. Factors regarding safety and environment were included into the project at the later stages where the only alternative is to use add-on protective equipment. Accidents like Flixborough and Bophal contributed in arising stricter regulations and increase public concern about process safety. This situation has led plant designers to incorporate inherent safety techniques believing that process design could be improved so that as many opportunities to reduce risk as possible are identified and implemented. The present work reviews the actual situation of IS not only looking backwards but also trying to conclude what is needed in the near future. Furthermore a recompilation of the most important methodologies in this area is made to show the issues concerning each one as well as their improvement along the years. This paper describes the principles introduced by Kletz which are the bases of all the strategies formulated for the methodologies constructed until now and shows how the situation has evolved since the introduction of the IS philosophy. After that, quantitative and qualitative techniques available or under development, which can be consulted in the literature and can be applied in different stages of the design are reviewed. Finally, comparative analysis of the indexes is performed to select safer process alternatives for acrylic acid production.

To whom all correspondence should be addressed. Address: Cno. La Carrindanga Km 7, C.P. 8000, Bahia Blanca, Argentina E-mail: [email protected]

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

2. Inherent Safety Philosophy

2.1. Inherent Safety and User-Friendly Principles Traditional steps used in assessing risk are identify, prevent, control and mitigate. For many years designers have associated the term prevent to the addition of protective equipment. Nowadays, additional principles like avoid or reduce have been defined in order to include the idea of eliminating or reducing the hazards permanently. Principles could be separated regarding IS and friendlier plants (Table 1), although in a sense, they merge each other (Kletz, 1984).

Table 1. Inherent Safety and User-Friendly principles Principle Inherent Safety Intensification or Minimization Sustitution Attenuation or Moderation Limitation of Effects User-Friendly Simplification Avoid Domino Effects Making Incorrect Assembly Impossible Making Status Clear Tolerance of Misuse Ease of Control Computer Control Instructions and Other Procedures Life-Cycle Friendliness Passive Safety Inventory reduction Change flammable refrigerant Store at low pressure Operate the reactor in gaseous phase Design change can avoid the need for better instrumentation Leaks can be dispersed by natural ventilation if plants are built out-of-doors To prevent relief valves from being installed the wrong way round, their inlet and exit lines should have flanges of different sizes A storage tank made from translucent plastic is friendly because the level in it is never doubt Expansion loops in pipework are more tolerant of poor installation than bellows Control Engineers should be involved early in design so that the control system is based of physical principles The software could be made more scrutable Write and instruction that covers most of the circumstances that can arise and is short and simple enough to be read, understood When designing a plant it should be considered the problems of those who have to construct and demolish it as well as those who have to operate it Use of fire insulation because it does not have to be commissioned Example

2.2. Past, Present and Future IS has passed through a lot of changes in mind since its conception in the seventies until these days. The idea starts from many comments, conclusions and recommendations arisen from discussions about major industrial accidents. In this sense, Flixborough left as lesson learn the important concept of inventory reduction (Kletz, 1975). On the other hand, incidents like that occur in nitroglycerin manufacture, where a batch process was replaced by a continuous one, was an example of attenuation (Kletz, 1998). Bhopal, the worst industrial disaster ever occurred, became evident the necessity to reduce hazard intermediate products. Nowadays, information regarding accidents with their corresponding causes and lack of IS principles can be found in the literature (Kletz, 1998; Gupta 2003; Sanders 2003; Kletz, 2003a; Lees, 1996). Even though the potentiality of IS is undeniable, their principles were proven to be complex on its full implementation and that is the reason why industry has not widely implemented them on their risk assessment 2

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

procedures yet. The feasibility of implementing the IS principles in industrial firms is discussed on literature (Zwetsloot and Ashford, 2003) and at the moment, in actual practice IS seems to be more of a conceptual methodology rather than a codified procedure with a well established and understood framework for evaluation and implementation. Some key obstacles had led IS to be underutilized (Moore and Rogers, 2003) i.e. (1) the perception that IS is technically and economically impractical for all but new processes, (2) the lack of an infrastructure and methodologies that permit IS to be integrated into technical, economic and EHS design considerations, and (3) the lack of industry guidance on the how to conduct a study, particularly for existing plants. Kletz (1996) stated some actions needed, such as making time available for the systematic study of alternatives during the early stages of design, research chemists should look for inherently safer processes, and senior managers need to emphasize the importance of Inherently Safer design on daily procedures and so on. But each constraint has its explanations (from the procedures and scheduling of industry) because regarding time available, new plants are wanted soon and the phrase time is money is incorrectly used since even when the inherently safer design takes more time, costs from development to decommissioning are lower. Nowadays and seven years after, Kletz (2003) makes another summary about the situation of Inherently Safer Design where it is observed the necessity of explaining which is the advantage of implementing this approach in design and industry (like in this abstract). Again, the same constraints are found but the one about a methodology to quantify IS is changing. IS is now being considered by regulators and environmentalists as the first process safety strategy industry that should be used for reducing process risk from such events as terrorism and safety in general (Moore and Rogers, 2003). Besides all the features discussed above, there is another one to be implemented in the near future and is related to undergraduate education. It is really necessary to implement these concepts early in the subjects of a future chemical engineer in a way that in his or her future work regarding design IS will be part of the philosophy towards an inherently safer process.

3. Description of reviewed methodologies

Either extrinsic safety and inherent safety mainly represent the total safety of a plant. However, both aspects need to be properly developed in order to keep total risk into tolerable levels. Particularly, this paper focuses on the IS methodologies from their conception until the latest developments. Since 1993, the interest in IS has been growing up in the researching field and many indices regarding this approach have been improved. This new concept is applicable in all the stages of design, but mostly in the early ones because it is the time where changes can be made easily. Furthermore, it can be realized that the lack of information in the early stages is noteworthy and this explains why the first indices are so simple. But these older indices helped in the improvement of all the ones that are under development nowadays. Prototype Inherent Safety Index. This reaction-oriented index was first implemented in the assessment of IS of the chemical process routes in the methyl methacrylate manufacture (Edwards and Lawrence, 1993; Lawrence, 1996). 3

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

In Lawrence work (1996), a preliminary list of parameters was proposed for possible incorporation in the index, but only the ones namely inventory, temperature, pressure, reaction yield, flammability, explosiveness and toxicity were used. Each parameter is tabulated with a domain, which is divided to shield the scale for its given score, in some cases, based on known methodologies such as Dow and Mond indices. A main division between a chemical and process score is made, where this chosen group of parameters are distributed depending on their characteristics. The chemical score takes into consideration the properties of the chemicals involved in the step and the process score involves the reaction conditions. After calculating the chemical and process scores for a given step, they are summed to obtain the total index for the step. Finally, the prototype inherent safety index for the route is the result of summing the score of each step in the route. Paying special attention to the characteristics of the parameters involved it can be noticed that this kind of information is available in process literature and material safety data sheets. This means, the index is applicable in the first stages of design where only information of this type is available. Inherent Safety Index. This index can be applied to conceptual process design and the aim is to extent an existent computerized method (Hurme and Jrvelinen, 1995) for process pre-design to encompass safety aspects using as a case study the Acetic Acid production (Heikkil, 1999). The method has two stages. The first is a rule-based synthesis where safety rules are combined with process rules. The other one is a safety analysis of synthesized alternatives, which is done by calculating safety indices based on the safety properties of chemical processes and the structure and conditions of the process. This RuleBased system generates process alternatives, which are simulated to determine the material and heat balances and to estimate physical properties. IS aspects are expressed in terms of an Inherent Safety Index, which consists of two main categories, the Chemical Inherent Safety Index which considers the effect of the choice of raw materials and other chemicals and the Process Inherent Safety Index counting the effect of the type of process equipment and processing conditions. Among the sub-indices used in this index, some were introduced by Edwards and Lawrence (1993) and the new ones namely heat of side reaction, corrosiveness, chemical interaction, equipment safety and safety of process structure show the difference in application since they are taken into account aspects related to process flowsheet data. INSIDE Project and the INSET Toolkit. This Tool was developed in Europe between 1994 and 1997 by the INSIDE Project being the report released in 2001 (Mansfield, 2001). The aim of this project is to develop a systematic tool to increase awareness toward IS principles among engineers and chemists and to encourage the application of IS to process and plant design. The methodology can be applied through the life-cycle of a processing plant. The INSET Toolkit consists of thirty-one tools and methods divided into four stages, which correspond to the plant life-cycle. All decisions involved in this methodology are related to safety, health and environmental hazards (SHE).

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

The tools available allow the identification and evaluation of imminent hazards and the approach is enough flexible to be applied to plant optimization based on Safety Health and Environmental (SHE) features as well as economics. The INSET Toolkit has a great flexibility given that each group of Tools can be applied not only in the early stages where we have to choose the chemical route but also in later ones where we could have more derailed information about the process. Expert System for the Design of Inherently Safer Processes. This approach is divided into two parts that correspond to route selection and process flowsheet selection. It begins with the construction of a methodology and the posterior implementation in an expert system, which automates the methodology. Route Selection Stage. Palaniappan et. al. (2002a) present a new IS index for ranking of process routes and a graphical method for analyzing reaction networks. This work introduces a methodology that can be applied to the first stages of design namely product specification stage, route selection stage and preliminary route screening stage. Each of these need specific information (Douglas, 1988) such as physical and chemical properties that lately would be scored for the calculation of the index. The score assigned to a given property depends on the parameters adopted for the division of the domain. For example, the reactivity of a substance is divided according to its NFPA (National Fire Protection Agency) scoring. In this approach, a process route is compared based on an overall safety index (composed by an Overall Chemical Index and an Overall Reaction Index which both constitute the Overall Safety Index) along with three other supplementary indices that help in the final decision in cases where the overall index does not show the factor that makes a route hazardous (worst chemical, total chemical and worst reaction indices). Flowsheet Development Stage. In this stage of process design, Palaniappan et. al. (2002b) present a methodology for IS analysis which is implemented in an expert system called iSafe, which enables the designer to identify alternatives at early stages of design by performing IS analysis having the most important advantage of reducing time and effort. The idea of such an expert system and the direct relation between IS and environmental issues have led to other expert systems (Palaniappan, 2002c). Similar to reaction graphs used in chemistry evaluation (Palaniappan, 2002a), a process graph representation is used as a model of the flowsheet to show the cause and effect behavior of the materials in the process. This part presents the Inherent Safety Index for Flowsheets where flowsheets are ranked using indices that are based on the scoring scheme proposed by Heikkil (1996). Palaniappan work describes the heuristics for safety analysis knowledge in commonly used equipment such as reactors, separators, heat exchangers and storage tanks. iSafe expert system has been implemented in different case studies such as selection of routes for phenol manufacture (Palaniappan et. al. 2002a), flowsheet development for the acrylic acid process (Palaniappan, 2002b) and selection of inherently safer process routes for acetic acid manufacture (Palaniappan et. al., 2004). The Simple Graphical Method. This method, proposed by Gupta and Edwards (2003), also applies to chemical process routes evaluation and extracts the comments introduced by experts in Lawrence thesis work 5

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

(Lawrence, 1996) and trying to improve the existing procedure. The comment that converted the Prototype Safety Index in a graphical method is referred to the idea of dimensions of the parameters involved. For example, the addition of temperature (C) and pressure (atm) result in a quantity that shouldnt be easy to compare with other alternatives and it is proposed that all terms should have the same dimensions or be made dimensionless. Furthermore, the comments were made over the existing parameters. Integrated Inherent Safety Index. This index is called integrated because it is intended to consider the life cycle of the process with economic evaluation and hazard potential evaluation (Khan and Amyotte, 2004a; 2004b). It is divided in two sub indices, the Hazard Index which is a measure of the potential damage of the process after taking into account the process and hazard control measures and the Inherent Safety Potential Index which accounts for the applicability of the IS principles (or guidewords) to the process. Each sub-index is calculated using the different curves presented in the previously mentioned work, taking into account fire, explosion and acute toxicity parameters, air water and soul pollutants and factors related to the operation of the process. Furthermore, the introduction of guidewords in the decision-making procedure makes this index similar to those currently being applied in industry such as Hazop. These characteristics make this index useful for both chemical process route selection and flowsheet selection. The calculations are joined to a set of frameworks, which help in the indexing procedure. Fuzzy Logic-Based Inherent Safety Index. Gentile (2004) developed and overall index to be used in process simulation and process synthesis to generate inherently safer alternatives and to evaluate them in a systematic way. The final objective of this methodology is to make it useful in the application of IS to operating plants. In this new approach, it is expected to solve some of the problems and limitations described for traditional IS evaluation methods. It comes as a solution for the oversensitive, which appears in common indices given that they manage discrete functions (Gentile et. al. 2003). Moreover, in IS evaluation there are different types of uncertainties. A mathematical approach based on fuzzy logic and possibility theory allows the modeling of the subjectivity and uncertainty introduced by the information needed. The combination of numerical information, subjective evaluation and heuristic knowledge into a single rigorous mathematical approach allows taking into consideration information that wouldnt be useful in traditional statistic but helps this analysis to be more detailed. All concepts and operations related to fuzzy logic are then implemented based on IF-THEN-RULES which, in some way, traduce the heuristics.

4. Indices Comparison
4.1. Case Study Description To discuss the characteristics of the indices presented, the acrylic acid production case study firstly introduced by Palaniappan et. al. (2002b) is used. The simplicity of this case study, allows a clearer analysis and comparison of the different aspects of the chosen indices. There are two main process routes for acrylic acid manufacture: one-step and two-step routes. The one-step process involves one main reaction and two possible side reactions, being the main one a catalytic oxidation of 6

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

propylene in the vapor phase with a range of temperature and pressure of 190 - 310C and 3 5 atm respectively. The two possible side reactions produce carbon dioxide with water and acetic acid with water. The two-step process involves oxidation of propylene to produce acrolein as the first main step. Acrolein is then oxidized to give acrylic acid and water; there is also the possibility of side reactions producing carbon dioxide and water from acrolein or propylene. These are catalytic reactions and are carried out at high temperature (250 320 C) in the vapor phase. Among the reactants and products involved in the reactions, the most important in what regards to flammability and toxicity are propylene, acrolein and acrylic acid. According to the NFPA Hazard Identification System, propylene has a rating of 4 in flammability characteristics; acrolein is classified as 4, 3 and 3 in relation to health, flammability and reactivity respectively and acrylic acid has a rating of 3 for health. The average process inventory was taken as 50000 metric tones, assuming 8000 hours per year and one hour hold up. 4.2. Index calculation for the Acrylic Acid Processes Among the indices presented in the review section, the Prototype Inherent Safety Index (PIIS), the Inherent Safety Index (ISI) and the i-Safe Index were calculated to carry out the comparison. The value calculated for each index is shown in Table 2 in a similar way as it was presented by Rahman et. al. (2004). The ISI has two sub-indices related to equipment and process safety. The main advantage of this is related to the accuracy achieved in the index calculation when there is certainty in the data needed. But, the lack of flowsheet information could be a drawback since more expertise is required. A wider analysis in order to classify the last sub indices and the others used in the calculations were tabulated by Rahman according to the source of the information involved. Hence, it is assumed that information about equipment and process is not available yet and so a sub-index of two was used (for details of the characteristics concerning the value of the sub-index see Heikkil thesis work (1999)).

Table 2. Sub-indices values for the methods applied.

PROC main 11 side 11 side 12 main 21 main 22 side 21 side 22 Hr Re Int FET Co Invent Yield Temp Pressure Eq Pr Total

ISI iSa iSa ISI ISI PIIS iSa ISI ISI PIIS PIIS ISa ISI PIIS iSa ISI PIIS iSa ISI ISI ISI PIIS iSa 4 4 4 4 4 4 4 4 4 4 4 4 4 4 2 1 1 3 3 1 3 2 2 2 7 7 6 8 7 6 7 7 6 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 5 1 1 1 1 1 1 5 1 1 1 1 1 1 2 2 2 3 2 3 2 3 3 3 4 3 4 3 2 2 2 3 2 3 2 0 0 0 0 0 0 0 1 1 1 1 1 1 1 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2 2 2 2 2 2 2 21 18 20 21 13 15 20 12 14 26 25 20 25 18 17 13 17 23 22 15 22

1 12 11 12 1 12 11 12 1 6 6 6 1 12 11 12

i-Safe index was recalculated for posterior comparison finding some differences in the results and in the data used. This deviation in the results could be attributed to the lack of information in the case study reported in the literature, for example data about yield for the second main reaction in the two-step route must be assumed. 7

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

The Table 2 shows in bold the value of the index corresponding a given step and the indexes concerning the side reactions are included as a complement. The step result does not include the heat of side reaction value. For a given step, comparison between the three indices is not advisable given the different scales in each and so absolute values could not be taken. Instead, considering the same index, and ranking from less safe to safer, a relative comparison could be made without losing consistency. As it was expected, the three indices state that two-step process route is the most hazardous one.

Conclusions
The number of IS Evaluation tools has been growing up during the last decade. The reasons of this evidence, found in most of the publication reviewed, include not only the improvement but also the extension of the first indices introduced in this approach. The INSET Toolkit was the first attempt in inherent process safety evaluation, offering a group of stages analysis that can be applied in many of the design stages. Quantitatively, the first index proposed by Edwards and Lawrence (Prototype Inherent Safety Index) was used mainly for process route evaluation and has been revised by several experts helping in its improvement. To implement the improvement proposed by experts, Edwards and Gupta introduced a simple graphical method. Later, Heikkil et. al. improved Edwards work by including in its indexing procedure a set of two sub indices that account for process and equipment safety, which helps in a more detailed evaluation including IS analysis among different equipment in the same process. The scoring procedure introduced by Heikkil was then considered by Palaniappan et. al. proposing a new index and developing an expert system called i-Safe for its implementation in evaluation of process flowsheet alternatives. Khan and Amyotte constructed an Integrated Inherent Safety Index proposing a conceptual framework for safety and cost evaluation and since they understand that the information needed for the calculation is not so easy to find like in the other indices, fuzzy logic is proposed to overcome uncertainties. Furthermore, fuzzy logic was applied as a new hierarchical approach in IS evaluation by Gentile et. al. Comparing the indices used in the case study calculations each index has its main advantage depending on the information available. Among the advantages found in ISI and i-Safe indices, the accuracy is improved since the range for scoring is shorter. PIIS works well when the evaluation is among reaction steps but does not consider reaction hazards at all (e.g. it does not has a reactivity sub-index). Anyway, it is really easy to use given its straightforward nature since information can be taken from material safety data sheets. i-Safe is also an index easy to use. According to its sub-index composition it shows to be reaction oriented like PIIS. The main change introduced by i-Safe over PIIS is that it covers reaction hazards introducing penalties for heat of reaction and reactivity. On the other hand, ISI could be more time consuming because there are needed more details like a process diagram to determine for example, the type of equipment. The calculation of heat of reaction either takes time and factors involving chemical interaction and corrosion are more laborious to find. All the observations previously stated can be summarized in outstanding comments about accuracy, applicability and scope. Firstly, certainty in the data needed is related to the accuracy achieved in an index 8

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

calculation, but the lesser the information available the greater the expertise required. Secondly, regarding the comparison between the indices, it must be highlighted that indices are calculated from different basis, and then it is not properly to compare their absolute values. Instead, for keeping consistency purposes, only relative conclusions within the same index must be done. Finally, in spite of being a cost-efficient strategy, IS principles have not been widely used yet. Some reasons for this may be the fact that the industry is still more familiarized with conventional safety strategies and many of the proposed methodologies are not so simple to be implemented easily. Nevertheless, the lack of systematic tools to apply inherent safety principles is perhaps the most important cause. In this context, the development of strategies applicable in practice should be among the priority objectives in future works.

References
Dalzell, G. (2003). Is Operating Cost a Direct Measure of Inherent Safety?. American Institute of Chemical Engineers. Trans IChemE, Vol. 81B, pp. 414-417. Douglas, J. M. (1988). Conceptual Design of Chemical Processes. Mc-Graw Hill. Edwards, D. W. and Lawrence, D. (1993). Assessing the Inherent Safety of Chemical Process Routes: Is There a Relation Between Plant Costs and Inherent Safety?. American Institute of Chemical Engineers. Trans IChemE. Vol. 71B, pp. 252-258. Gentile, M. (2004). Development of a Hierarchical Fuzzy Model for The Evaluation of Inherent Safety. Ph. D. Thesis. Texas A&M University, Texas. Gentile, M., Rogers, W. J., Mannan, M. S. (2003). Development of a Fuzzy Logic-Based Inherent Safety index. Institution of Chemical Engineers. Trans IChemE, Vol. 81B, pp. 444-456. Gupta, J. P. (2003). The Bophal Gas Tragedy: Could It Have Happened In a Developed Country?. Journal of Loss Prevention in the Process Industries, 15, pp. 1-4. Gupta, J. P., Edwards, D. W. (2003). A Simple Graphical Method for Measuring Inherent Safety. Journal of Hazardous Materials, 104, pp. 15-30. Heikkil, A-M. (1999). Inherent Safety in Process Plant Design: An Index Based Approach. Ph. D. Thesis. VTT Publication Number 284, Helsinki University of Technology, Espoo, Finland. Heikkil, A-M., Hurme, M., Jrvelinen, M. (1996). Safety Considerations in Process Synthesis. Computers Chem, Eng., Vol. 20, pp. S115-S120. Hurme, M, Jrvelinen, M. (1995). Combined Process Synthesis and Simulation System for Feasibility Studies. Computers Chem, Eng. Vol 19, pp. S663-S668. Khan, F. I., Amyotte, P. R. (2004a). Integrated Inherent Safety Index (I2SI): A Tool for Inherent Safety Evaluation. Process Safety Progress, Vol. 23, No. 2, June, pp. 136-148. Khan, F. I., Amyotte, P. R. (2004b). I2S: A Comprehensive Quantitative Tool for Inherent Safety and Cost Evaluation. International Conference on the 20th Anniversary of the Bophal Gas Tragedy. Indian Institute of Technology, Kanpur, India. December. Khan, F. I., Sadiq, R., Amyotte, P.R. (2003). Evaluation of Available Indices for Inherently Safer Design Options. Process Safety Progress, Vol. 22, No. 2, June, pp. 83-97. Kletz, T. A. (1975). Loss Prevention, Vol. 9, pp. 106. Reprinted, with minor changes in Chemical Engineering, Vol. 83, No. 8, p. 124 (April 12, 1976). Kletz, T. A. (1996). Inherently Safer Design: The Growth of an Idea. Process Safety Progress, Vol. 15, No. 1, Spring, pp. 5-8.

2nd Mercosur Congress on Chemical Engineering 4 Mercosur Congress on Process Systems Engineering
th

Kletz, T. A. (1998). Process Plants: Handbook for Inherently Safer Design. Taylor & Francis. Kletz, T. A. (2003a). Still Going Wrong!. Gulf Professional Publishing. Kletz, T. A. (2003b). Inherently safer Design Its Scope and Future. Institution of Chemical Engineers. Trans IChemE, Vol. 81B, pp. 401-405. Lawrence, D. (1996). Quantifying Inherent Safety of Chemical Process Routes. Ph. D. Thesis, Loughborough University, Loughborough, U. K. Lees, F. P. (1996a). Loss Prevention in the Process Industries: Hazards Identification and Control. 2nd ed. Vol 3. Lees, F. P. (1996b). Loss Prevention in the Process Industries: Hazards Identification and Control. 2nd ed. Vol 1. Mansfield, D. (2001). The INSIDE Project Inherent SHE in Design. Available at: http//www.aeat-safety-andrisk.com/html/inset.html. Moore, D. A., Rogers, J. E. L. (2003). The Challenge of Inherent Safety. CCPS International Conference Proceedings. Palaniappan, C., Srinivasan, R., Tan, R. (2002a). Expert System for the Design of Inherently Safer Processes. 1. Route Selection Stage. Ind. Eng. Chem. Res., Vol. 41, pp. 6698-6710. Palaniappan, C., Srinivasan, R., Tan, R. (2002b). Expert System for the Design of Inherently Safer Processes. 2. Flowsheet Development Stage. Ind. Eng. Chem. Res., Vol. 41, pp. 6711-6722. Palaniappan, C., Srinivasan, R., Halim, I. (2002c). A Material-Centric Methodology for Developing Inherently Safer Environmentally Benign Processes. Computers and Chemical Engineering, Vol. 26, pp. 757-774. Palaniappan, C., Srinivasan, R., Tan, R. (2004). Selection of Inherently Safer Process Routes: A Case Study. Chemical Engineering and Processing, Vol. 43, pp. 647-653. Rahman, M., Heikkil A-M., Hurme, M. (2004). Comparison of Inherent Safety Indices in Process Concept Evaluation. International Conference on the 20th Anniversary of the Bophal Gas Tragedy. Indian Institute of Technology, Kanpur, India. December. Sanders, R. E. (2003). Designs That Lacked Inherent Safety: Case Histories. Journal of Hazardous Materials, Vol. 104, pp. 149-161. Zwetsloot, G. I. J. M., Ashford, N. A. (2003). The Feasibility of Ecouraging Inherently Safer Production in Industrial Firms. Safety Science. Vol. 41, pp. 219-240.

Acknowledgments
The authors gratefully acknowledge Dr. David Edwards and Dr. Rajagopalan Srinivasan for the information supplied. The authors acknowledge Universidad Nacional del Sur, CONICET and ANPCyT for financial support.

10